Welcome to my project exploring U.S. and global consumers—their demographics, shopping habits, and preferences! Using interactive data visualizations, I discovered patterns in customer spending, their product preferences, device usage, discount sensitivity, and brand loyalty. These insights could help marketers and businesses to segment high-value customers and design more targeted marketing strategies. You will also see how tailored marketing by geography and demographics, along with seasonal and product-based trends, can improve marketing effectiveness and possibly reduce marketing costs. Finally, we examine how brand loyalty is shaped by customer’s purchase intent, digital engagement, and consumer profiles, which are insights that marketing specialists can learn from to build long-term customer connections.
My motivation for creating this project is my interest in marketing and particularly in understanding how people make decisions and what drives their behavior as consumers. With a background in both marketing and statistics, I wanted to tell a story through data visualization, which I believe can be more intuitive and impactful than complex models or algorithms, especially for audiences who prefer minimal technical jargon. My hope is that this project offers marketing specialists, analysts, and consultants helpful, data-driven insights on how to better meet customer needs, raise product awareness, and potentially discover new opportunities to connect with and retain loyal customers.
In this project, I used two consumer shopping datasets from Kaggle, both based on consumer studies and market research. The first dataset focuses on U.S. consumers, including age, location, and online shopping behaviors, last updated five months ago. The second, the Ecommerce Consumer Behavior Analysis dataset, includes global e-commerce trends, consumer demographics, preferences, and behavioral patterns.
Let’s start with the first visualization, a U.S. map showing the average purchase amount by state. States shaded in darker colors have higher spending whereas the lighter ones have lower spending. You can hover over each state to see the state and its average customer purchase amount in USD. As you might notice, Pennsylvania and Arizona stand out with the highest average spending. This raises interesting questions for market researchers: Why are customers in these states spending more? Could it be due to socioeconomic factors or local shopping trends? For online retailers, this insight can help optimize inventory and target ads in high-spending states to better meet demand.
Now that we’ve seen the big picture of how much customers in the United States and each state spends, let’s dig into some of the more detailed trends in the dataset.
Now, let’s explore customer purchasing trends through customer segmentation, a common method widely used in marketing to break down a broad customer base into smaller groups based on shared characteristics. Visualizing the segments and the variable in interest (in this case total purchase contribution, equivalent to total revenue contribution, from retailer’s perspective) makes it much easier to understand which segments are driving the most value. In this section, we’ll group customers by age and gender, which are two key factors that often influence shopping trends and behaviors.
This treemap breaks down total purchase (233,081 dollars) by gender and age group. There are six age categories (≤20 to 60+) and two gender levels (male, female). As shown, male customers contributed about twice as much as female customers, 157,890 vs. 75,191. Within each gender group, the top contributing age groups are 51–60 and 60+, suggesting that older customers tend to spend more, possibly due to greater savings or more free time to shop online after years of hard work! This interactive treemap lets you click through each box to explore segment-specific values. For example, you may click on the smallest box, females under 20, which presents the lowest revenue contribution. Now, please interact with this treemap to explore revenue contribution by customer gender and age.
After segmenting customers by gender and age, we can now look more into differences in shopping preferences across different age groups. Visualization 3 shows a grouped bar chart of total purchase amount by age group across four product categories, accessories, clothing, footwear, and outerwear. Each age group is color-coded, and you can hover over the bars to explore how different age groups spend across categories.
Clothing clearly drives the most overall revenue, followed by accessories. Within each category, we can see distinct trends across the age groups. 60+ customers spend the most on clothing and accessories, 51–60 spend themost in footwear purchases, and 41–50 spend the most in outerwear. These insights suggest that marketers should tailor marketing campaigns and personalize advertising messages based on both product type and age group to maximize effectiveness of marketing efforts and save marketing budget. Furthermore, you may also see that clearly, customers from the ≤20 age group contribute the least across all age groups across all categories. From many marketing studies, this is not surprising given that customers younger than 21 are still financially dependent on their parents. Interestingly, consumers from the 21–40 age group are also not spending much as well, possibly due to their early career and financial obligations like paying for their NYC apartment rents. While this is an assumption, together with more careful investigations and market research, retailers might be able to discover some unique channels to reach these younger customer groups. However, together with our earlier observation, older age groups are the highest spenders who create the most value to the business. Last but not least, clothing and accessories categories show high revenues earned across all age groups. This may reflect the impact of online advertising and influencer marketing since these items are easier to market and buy online compared to footwear, which often requires trying on in person due to fit and quality expectations. But again, we would need a bit more information to explain the differences and it would be an interesting topic to study for businesses.
Building on the previous observation of the clothing product sales, this lollipop chart shows the top clothing items based on total revenue/total purchase amount. Each line ends with a solid circle which highlights the total revenue for that item and allows you to make comparisons across the clothing items.
We can see that blouses, shirts, dresses, and pants generate the most revenue, which signal strong consumer demand. For businesses and marketing specialists, these items may present opportunities for more promotions or seasonal sales to increase revenues. Spring and fall trend drops or end-of-season sales could all further boost profitability, especially when paired with influencer marketing and social media ads, which continue to play important roles in driving consumer preferences. The next animated visualization will highlight the most popular item colors purchased across season, for example, spring florals or warm golden fall tones, that often influence purchases in categories like clothing and accessories.
This animated bar plot shows the number of items purchased by color across each season. The clear shifts in color popularity suggest that color plays a role in driving consumer preferences within the apparel category! As you see, item colors change noticeably by season. For example, warm hues like yellow, olive, and orange are more popular in fall, while pastels and cool tones like pink, cyan, and silver are more popular in warmer months like spring and summer. These trends suggest that seasonal color palettes influence purchasing behavior. Therefore, marketers could consider aligning product designs and designing marketing campaigns using seasonal color palettes to increase aesthetic appeal and marketing visuals. However, one thing to note is that colors like Black and Gray, though not always among the most popular seasonal colors, tend to have high purchase counts across all seasons. This is potentially due to their ease of pairing with other colors. Because of this, marketers may focus less on analyzing these colors and instead focus their efforts on more dynamic seasonal color trends.
Together with your findings and insights gained from these four visualizations, I hope by this point, you could have noticed some trends over the shopping habits and the profitability of different consumer segments for the overall platform and for specific product categories. Whether your brand is trying to expand the customer segment, boost customer engagement with high-value customers, or design personalized marketing strategies and campaigns, these data-driven insights can help you make wiser decisions! With the insights from these visualizations, I hope you’ve started to notice patterns in U.S. consumer shopping habits and the profitability of different customer segments both across the entire platform and within specific product categories. Whether your goal is to expand customer reach, strengthen engagement with high-value segments, or design more personalized marketing strategies, these data-driven insights can help you make wiser decisions!
Now, let’s switch gears and turn our focus to global consumers. Together, we have already explored U.S. consumer behavior and identified how active and profitable each age group is across various factors therefore now, it is time to see how these patterns vary on a broader, international scale.
To kick off our global consumer analysis, we will again begin with a map. This interactive world map allows you to explore consumer behavior patterns across various geographic regions. Each blue dot represents a unique location, and by clicking on it, you will get to see local insights under the “Location Insight” panel, including their social media influence, ad engagement, purchase intent, and more. Because this dataset includes many unfamiliar or smaller cities, a reference map with continents labeled is also provided to help guide your exploration. For example, by hovering over the North America region and clicking on Philadelphia, you’ll be able to see its total purchase amount and a detailed breakdown of that location’s consumer response to marketing efforts. These insights can help marketers better understand how consumers across different regions respond to marketing efforts, combined with our earlier findings on U.S. consumers.
https://tmu7se-2025.shinyapps.io/project-viz-5/
More importantly, this visualization can help brands evaluate whether it’s worthwhile to launch targeted campaigns in specific areas. By examining metrics like discount sensitivity and purchase intent, marketers can decide whether to offer promotions or incentives in particular regions. Similarly, insights into loyalty program participation can guide decisions around retention strategies.
Building on the geographic insights from the interactive world map, this next visualization provides you more details about global consumer behavior through specific personal demographics. Feel free to explore this interactive Shiny App to see how factors like education level, age, and income are associated with different shopping behaviors. By selecting multiple demographic filters, you can not only study specific customer segmentation (similar to our exploration in part 1) that you are interested in, but also participate in generating visualizations with me that summarize customers’ loyalty program status with age distribution, and examine their discount sensitivity and preferred purchasing channels. Using this interactive app, marketing analysts and specialists can explore consumer behavior across different groups, by age, income level, occupation, and more.
https://tmu7se-2025.shinyapps.io/project-viz-6/
The age distribution histogram above reveals how loyalty program participation varies across age groups. This can be interpreted in relation to earlier insights from Part 1 on U.S. consumers, helping us better understand which age segments are more engaged in brand loyalty programs. The donut chart below displays the distribution of preferred purchase channels, tablet, desktop, and smartphone, among consumers. You can filter by age range and other inputs to see how device preferences shift across different age groups. For certain demographics, they might be more frequent smartphone users compared to other groups. For businesses, they could then make targeted strategy to communicate with their customers through their preferred type of technology. Lastly, the discount sensitivity waffle chart shows the proportion of customers within the selected segment who are highly, moderately, or less sensitive to discounts. This can help marketers decide whether offering promotions, despite smaller profit margins, would be effective in increasing sales among that customer group as well as avoiding unnecessary marketing spend wisely.
As we know more about consumers’ shopping behaviors, we can now look into the journey of customer experience, specifically what factors drive customer satisfaction, and how these factors are interconnected. The jitter plot shows the relationship between customer satisfaction and product ratings, split into two panels by income group (high and middle), with each dot representing a customer and colored by their level of brand loyalty (low, medium, or high). Although not clearly presented, we can still roughly see that higher product ratings generally correspond to greater satisfaction, as seen in the dense clusters of points in the upper-right areas of each panel. Customers with higher brand loyalty (purple) tend to be more satisfied, especially among high-income individuals, where satisfaction is both higher and more consistent. This suggests a more positive overall experience for loyal, high-income customers. In contrast, middle-income customers display more scattered satisfaction across all product ratings and have fewer highly loyal members. This suggests that their satisfaction may depend more on factors like price sensitivity (comparing middle vs. high income customers). Even when product ratings are similar, the differences in satisfaction shows how income and loyalty together could contribute to the overall customer experience. This kind of visualization is extremely useful if you want to examine how satisfaction varies with product ratings and loyalty levels, and how these relationships differ across income segments. For marketing analysts and customer insight specialists, this tool is valuable for segmenting customers and identifying opportunities for more targeted promotions. By observing association between product ratings and customer satisfaction, teams can refine loyalty programs, personalize messaging, and better understand customer experiences. The next visualization focuses on exploring brand loyalty specifically.
Last but not least, we will wrap up this project by exploring one of the most critical drivers of long-term business success which is brand loyalty. This Shiny App with an output sunburst visualization allows you to filter customer data by demographics and product category to discover how many customers fall into each behavioral and ultimately loyalty status based on filters they choose.
https://tmu7se-2025.shinyapps.io/project-viz-8/
The sunburst chart presents a three-level hierarchical view of consumer behavior. Level 1 (center) shows purchase intent (e.g., impulsive, wants-based, need-based, or planned). This motivation can vary by age, income level, and more. For instance, younger consumers may be more driven by wants and might be more impulsive. Level 2, social media engagement, shows how connected and responsive different consumer groups are to digital marketing depending on their purchase intent. For example, you could explore whether higher activity on social media and more likely to engage with influencer-driven content(more engagement on digital advertising) could be associated with more impulsive buys. Level 3 reflects brand loyalty levels, ranging from super loyal to not loyal and nested within each purchase intent and engagement level. This allows you to explore questions like: Are high social media engagement consumers also more brand loyal? Consider the concept of “brand-first” vs. “product-first” buying behavior as you interact with the app. Do loyal Lululemon shoppers represent strong brand-first behavior, while toothpaste buyers prioritize function over brand? Observing these patterns is especially important when evaluating whether consumers are buying “product before brand” (e.g., toothpaste) or “brand before product” (e.g., Lululemon leggings). For example, in the latter case, a strong brand identity drives loyalty even at premium prices. To launch new products and market them more effectively, business marketers can use this type of sunburst chart to design a customer roadmap from building awareness to guiding consumers into the loyalty loop, ensuring long-term success.
The retail market is becoming increasingly competitive, as companies invest more in understanding customer behavior and leveraging technology and digital platforms to engage consumers through targeted marketing. Through this data visualization project, I hope you’ve gained useful insights into modern consumer behavior and trends that can help your business grow—or at the very least, discovered some intuitive and effective visualization techniques to guide your team’s next marketing campaign or product innovation.